JP2013033726A - Abnormality detection device and abnormality detection method - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide an abnormality detection device and an abnormality detection method that can improve determination precision of an abnormal discharge.SOLUTION: There is provided an abnormality detection device 70 including a monitor part 72 which monitors operation from the start of detachment of a wafer after plasma processing which is loaded in a processing chamber to opening of a transfer gate valve, and specifies the operation as operation in wafer detachment; an acquisition part 73 which acquires a high-frequency signal of at least one of a progressive wave and reflected wave output from a directional coupler provided between a high-frequency power supply which applies high-frequency electric power into the processing chamber and an impedance matching circuit or between a lower electrode functioning as a mount table where a workpiece is mounted and the impedance matching circuit during operation in the specified wafer detachment operation; an analysis part 74 which analyzes a waveform pattern of the acquired high-frequency signal; and an abnormality determination part 75 which determines whether the abnormal discharge has occurred on the basis of an analysis result of the waveform pattern of the high-frequency signal.

Description

  The present invention relates to an abnormality detection apparatus and an abnormality detection method for detecting abnormal discharge in a plasma processing apparatus.

  When fine processing using plasma is performed on an object to be processed such as a semiconductor wafer (hereinafter referred to as a wafer) or a substrate, a gas is introduced into a processing chamber in which plasma processing is performed, and high-frequency power is applied to introduce the introduced gas. Is decomposed to generate plasma, and the plasma is used to process the object to be processed.

  In a processing chamber where plasma is generated, a high-frequency electric field concentrates due to various factors, and abnormal discharge such as arc discharge may occur. Abnormal discharge leaves a discharge mark on the surface of the object to be processed or burns out components disposed in the processing chamber. In addition, the reaction product or the like attached to the components in the processing chamber is peeled off, which causes generation of particles.

  For this reason, abnormal discharge in the processing chamber is detected at an early stage, and when abnormal discharge is observed, appropriate action is taken such as immediately stopping the operation of the plasma processing apparatus, and damage or particles on the object to be processed or various parts It is necessary to prevent the occurrence of

  Therefore, conventionally, a method for detecting abnormal discharge at an early stage has been proposed. As an example, there is a method in which a test process is provided after wafer processing and a discharge trace is found by visual inspection or the like in the test process. However, according to this method, there is a considerable waiting time until the test process after wafer processing, and plasma processing is performed on unprocessed wafers one after another. As a result, even if a wafer defect is found in the test process, it takes a certain amount of time to stop the plasma processing apparatus in which abnormal discharge has occurred. A defective chip will be produced.

  As another method for early detection of abnormal discharge, it has also been proposed to detect abnormal discharge from the waveform pattern of the AE signal using an AE (Acoustic Emission) sensor (see, for example, Patent Document 1). .

JP 2011-14608 A

  However, since the AE sensor has good sensitivity, the AE signal detected by the AE sensor is not only a signal due to abnormal plasma discharge, but also a machine due to opening / closing of the transfer gate valve of the plasma processing apparatus and vertical movement of the transfer pin. It contains a lot of noise such as mechanical vibration. Therefore, there is a concern that the detection accuracy of abnormal discharge of the plasma processing apparatus may be reduced due to noise included in the AE signal.

  In view of the above problems, an object of the present invention is to provide an abnormality detection device and an abnormality detection method capable of improving the accuracy of determining abnormal discharge.

  In order to solve the above-described problems, according to an aspect of the present invention, a high-frequency power source matching unit that applies high-frequency power to a processing chamber that plasma-processes an object to be processed and a mounting table on which the object to be processed is mounted. An acquisition unit that acquires a high-frequency signal output from an RF sensor provided between the lower electrode and an AE signal output from an AE sensor for detecting acoustic emission generated in the processing chamber; An analysis unit for analyzing the waveform pattern of the high frequency signal and the waveform pattern of the AE signal, and an abnormality determination for determining the presence or absence of abnormal discharge based on the analysis result of the waveform pattern of the high frequency signal and the analysis result of the waveform pattern of the AE signal And an abnormality detecting device characterized by comprising a unit.

  The analysis unit extracts a maximum amplitude value of the high frequency signal and a maximum amplitude value of the AE signal during operation when the object is detached based on the waveform pattern of the high frequency signal and the waveform pattern of the AE signal. The abnormality determination unit may determine the presence or absence of abnormal discharge by comparing the maximum amplitude of the high-frequency signal with a first threshold and comparing the maximum amplitude of the AE signal with a second threshold. .

  The RF sensor may be a directional coupler, an RF probe, or a current probe.

  According to another aspect of the present invention, the operation of the object to be processed placed in the processing chamber is monitored from the start of desorption of the object to be processed after the plasma processing until the transfer gate valve is opened, and the operation is performed. A monitoring unit that is specified as an operation at the time of detachment of the object to be processed, and a high-frequency power source that applies high-frequency power to the processing chamber during the operation at the time of detachment of the specified object and a matching unit, or an object to be processed An acquisition unit for acquiring a high-frequency signal of at least one of a traveling wave and a reflected wave output from a directional coupler provided between the lower electrode functioning as a mounting table and a matching unit; Provided is an abnormality detection device comprising: an analysis unit that analyzes a waveform pattern of an acquired high-frequency signal; and an abnormality determination unit that determines presence or absence of abnormal discharge based on an analysis result of the waveform pattern of the high-frequency signal Is done.

  The acquisition unit acquires an AE signal output from an AE sensor for detecting acoustic emission (AE) generated in the processing chamber, and the analysis unit analyzes a waveform pattern of the acquired AE signal. The abnormality determination unit may determine the presence or absence of abnormal discharge based on the analysis result of the waveform pattern of the high-frequency signal and the waveform pattern of the AE signal.

  The analysis unit extracts a maximum amplitude value of the high frequency signal and a maximum amplitude value of the AE signal during operation when the object is detached based on the waveform pattern of the high frequency signal and the waveform pattern of the AE signal. The abnormality determination unit may determine the presence or absence of abnormal discharge by comparing the maximum amplitude of the high-frequency signal with a first threshold and comparing the maximum amplitude of the AE signal with a second threshold. .

    The abnormality determination unit may determine the presence or absence of abnormal discharge by comparing the generation time of the maximum amplitude of the high-frequency signal with the generation time of the maximum amplitude of the AE signal.

  The analysis unit may perform frequency analysis on the waveform pattern of the AE signal, remove noise from the frequency-analyzed data using a desired noise removal filter, and then analyze the data from which the noise has been removed.

  The AE sensor may be attached to a power supply rod that supplies high-frequency power to a lower electrode that also functions as a mounting table on which an object to be processed is mounted.

  The sampling of the high frequency signal may be performed every 1 μsec to 5 μsec.

  The sampling of the AE signal may be performed every 1 μsec to 1 msec.

  The monitoring unit may determine that when the output of the DC high voltage power applied to the electrode of the electrostatic chuck is turned off, or when the DC high voltage power is reversely applied, the start of desorption of the object to be processed after the plasma processing is performed. Good.

  In order to solve the above-described problem, according to another aspect of the present invention, a high-frequency power source matching unit that applies high-frequency power to a processing chamber that plasma-processes an object to be processed and a mounting table on which the object to be processed is mounted A high-frequency signal output from an RF sensor provided between the lower electrode and the AE signal output from an AE sensor for detecting acoustic emission generated in the processing chamber; and Analyzing the waveform pattern of the high frequency signal and the waveform pattern of the AE signal; determining the presence or absence of abnormal discharge based on the analysis result of the waveform pattern of the high frequency signal and the analysis result of the waveform pattern of the AE signal; The abnormality detection method characterized by including is provided.

  Further, according to another aspect of the present invention, the object to be processed placed in the processing chamber is monitored from the start of desorption of the object to be processed after the plasma processing until the transfer gate valve is opened, A step of specifying an operation as an operation at the time of detachment of the object to be processed, and a high-frequency power source that applies high-frequency power to the processing chamber during the specified operation of detaching the object to be processed and a matching unit Obtaining a high-frequency signal of at least one of a traveling wave and a reflected wave output from a directional coupler provided between the lower electrode functioning as a mounting table on which a body is mounted and the matching unit; There is provided an abnormality detection method comprising: analyzing a waveform pattern of an acquired high-frequency signal; and determining whether or not there is abnormal discharge based on an analysis result of the waveform pattern of the high-frequency signal. That.

  Furthermore, according to another aspect of the present invention, there is provided a plasma processing apparatus, comprising: a processing chamber for processing a substrate; plasma generating means for generating plasma in the processing chamber; and a power supply unit for the plasma generating means, An abnormal discharge detecting device for detecting abnormalities in plasma, wherein the abnormal discharge detecting device mounts a matching device of a high-frequency power source for applying high-frequency power and a processing object in a processing chamber for plasma-processing the processing object. An acquisition unit for acquiring a high-frequency signal output from an RF sensor provided between the lower electrode functioning as a mounting table and an AE signal output from an AE sensor for detecting acoustic emission generated in the processing chamber An analysis unit for analyzing the waveform pattern of the acquired high-frequency signal and the waveform pattern of the AE signal, and a waveform pattern of the high-frequency signal And having the analysis results and the abnormality determining unit for determining the presence or absence of an analysis result based abnormal discharge waveform pattern of the AE signal, the plasma processing apparatus is provided.

  As described above, according to the present invention, it is possible to provide an abnormality detection device and an abnormality detection method that can improve the accuracy of determination of abnormal discharge.

It is a longitudinal cross-sectional view of the etching processing apparatus which concerns on one Embodiment of this invention. It is a functional lineblock diagram of a control device for an etching processing device concerning one embodiment. It is the flowchart which showed the data acquisition process which concerns on one Embodiment. It is a figure for demonstrating an example of the abnormal waveform which arises with the etching processing apparatus which concerns on one Embodiment. It is the flowchart which showed the abnormality detection process which concerns on one Embodiment. It is the figure which showed the waveform pattern of the high frequency signal which concerns on one Embodiment for every wafer. It is the figure which showed the maximum amplitude of the high frequency signal which concerns on one Embodiment for every wafer. It is the figure which showed the waveform pattern of the AE signal which concerns on one Embodiment for every wafer. It is the figure which showed the waveform pattern after the frequency analysis of the AE signal which concerns on one Embodiment for every wafer. It is the figure which showed the maximum amplitude of the AE signal which concerns on one Embodiment for every wafer. It is the figure which showed the shift | offset | difference of the generation time of the maximum amplitude of the high frequency signal and AE signal which concerns on one Embodiment. It is a longitudinal cross-sectional view of the etching processing apparatus which concerns on other embodiment. It is the figure which showed the waveform pattern of the high frequency signal by the RF probe which concerns on one Embodiment, and the waveform pattern of AE signal. It is the flowchart which showed the abnormality detection process which concerns on a modification.

  Exemplary embodiments of the present invention will be described below in detail with reference to the accompanying drawings. In addition, in this specification and drawing, about the component which has the substantially same function structure, duplication description is abbreviate | omitted by attaching | subjecting the same code | symbol.

[Device configuration]
First, an etching processing apparatus is taken as an example of a plasma processing apparatus according to an embodiment of the present invention, and the configuration of the apparatus will be described with reference to FIG. FIG. 1 is a longitudinal sectional view of an etching processing apparatus according to an embodiment of the present invention.

  The etching processing apparatus 10 is a capacitively coupled etching apparatus in which electrode plates are arranged in parallel to face each other. The etching processing apparatus 10 has a processing chamber C made of, for example, aluminum whose surface is anodized and formed into a cylindrical shape. The processing chamber C is grounded. A substantially cylindrical susceptor support 11 for mounting the wafer W is provided on the bottom of the processing chamber C via an insulating plate (not shown) such as ceramics. On the susceptor support 11, a mounting table 12 that also serves as the lower electrode DE is provided.

  In order to fix the wafer W on the lower electrode DE, an electrostatic chuck 13 is provided on the lower electrode DE. The electrode 13a of the electrostatic chuck 13 is formed of, for example, a polyimide thin film and is embedded in the lower electrode DE. When a DC voltage output from the DC voltage source 14 is applied to the electrode 13a, the wafer W placed on the surface of the electrode 13a is attracted and held on the lower electrode DE by electrostatic attraction force.

  A gas passage (not shown) for supplying a heat transfer medium (for example, He gas) is formed on the back surface of the wafer W in the susceptor support base 11, the mounting base 12, and the electrostatic chuck 13. Thus, the cold heat of the mounting table 12 is transmitted to the wafer W so that the wafer W is maintained at a predetermined temperature.

  An upper electrode UE is provided above the mounting table 12 so as to face the lower electrode DE. The upper electrode UE is supported on the ceiling of the processing chamber C. The upper electrode UE includes an upper electrode plate 31 and an electrode support 32 made of a conductive material that supports the upper electrode plate 31.

  An etching gas or the like for plasma etching is supplied to the processing chamber C from a gas path (not shown). An exhaust device 35 is connected to the bottom of the processing chamber C via an exhaust pipe 34. The exhaust device 35 includes a vacuum pump such as a turbo molecular pump, and evacuates the processing chamber C to a predetermined reduced pressure atmosphere. A transfer gate valve 36 is provided on the side wall of the processing chamber C, and the wafer W is transferred between adjacent transfer chambers (not shown) by opening and closing the transfer gate valve 36.

  The etching processing apparatus 10 supplies high-frequency power of two upper and lower frequencies. A first high frequency power supply 40 is connected to the upper electrode UE via a matching unit 41. The first high frequency power supply 40 has a frequency (RF) in the range of 27 to 150 MHz, for example. By supplying high-frequency power between the upper electrode and the lower electrode while supplying the etching gas, a plasma of a preferable etching gas can be formed in the processing chamber C.

  A second high frequency power supply 50 is connected to the mounting table 12 as the lower electrode DE via a matching unit 51. The matching unit 51 and the lower electrode DE are connected by a power supply rod 52. A directional coupler 60 is provided between the second high frequency power supply 50 and the matching unit 51, and is connected by coaxial cables 53a and 53b. The second high frequency power supply 50 applies bias high frequency power having a frequency lower than that of the first high frequency power supply 40 to the lower electrode DE. Thereby, an appropriate ion action can be given to the wafer W without damaging it. The frequency of the second high frequency power supply 50 is preferably in the range of 1 to 20 MHz, for example. In addition, the directional coupler 60 may be provided between the lower electrode DE that also functions as a mounting table on which the wafer W is mounted and the matching unit 51.

  With this configuration, the etching gas introduced into the processing chamber C is turned into plasma by the high-frequency power output from the first high-frequency power source 40, and a desired etching process is performed on the wafer W on the mounting table using the generated plasma. Is given.

[Abnormal discharge detection]
Next, detection of abnormal discharge in the etching processing apparatus 10 will be described.

  During the etching process performed by the etching apparatus 10, high-frequency power is applied to the wafer W from the second high-frequency power supply 50 through the coaxial cable 53 a, the directional coupler 60, the coaxial cable 53 b, the matching unit 51, and the power supply rod 52. In this case, abnormal discharge (for example, arc discharge) may occur in the processing chamber C. At that time, the impedance of the plasma is disturbed, and a reflected wave BtmPr from the matching unit 51 to the second high frequency power supply 50 side is generated. The directional coupler 60 detects the reflected wave BtmPr from the matching unit 51 toward the second high frequency power supply 50. The directional coupler 60 can also detect a traveling wave BtmPf that travels from the second high frequency power supply 50 to the matching unit 51. However, the directional coupler 60 may detect at least one of the traveling wave BtmPf and the reflected wave BtmPr. Hereinafter, at least one of the traveling wave BtmPf and the reflected wave BtmPr detected by the directional coupler 60 is referred to as a high frequency signal.

  The AE sensor 61 is attached to the ground line of the power feed rod 52 with an adhesive or the like. The AE sensor 61 detects AE (Acoustic Emission) resulting from energy emission at the time of abnormal plasma discharge. In the present embodiment, the AE sensor 61 is attached to the atmospheric space outside the processing chamber C at a position as close as possible to the wafer W in order to detect abnormal discharge generated on the wafer. The AE sensor 61 may be attached above the wafer W. However, it is preferable to mount the AE sensor 61 below the wafer W because it is easy to detect abnormal discharge on the wafer W.

  Since the AE sensor 61 has high sensitivity, the AE signal detected by the AE sensor 61 is not only a signal due to abnormal plasma discharge but also the opening / closing of the transfer gate valve 36 of the plasma processing apparatus 10 and the mounted wafer W. It contains a lot of noise such as mechanical vibration caused by raising and lowering of a pin (not shown) for transporting. Therefore, there is a concern that the detection accuracy of abnormal plasma discharge may be reduced due to noise included in the AE signal.

  Therefore, in the detection of abnormal discharge according to the present embodiment, the directional coupler 60 and the AE sensor 61 are attached to the etching processing apparatus 10, the high frequency signal detected by the directional coupler 60 and the AE detected by the AE sensor 61. High-speed sampling with the signal. Hereinafter, the sampled AE signal is also indicated as “BtmAE”. Further, the sampled high frequency signal of the reflected wave is also displayed as “BtmPr”, and the sampled high frequency signal of the traveling wave is also displayed as “BtmPf”.

  The abnormality detection device 70 acquires the sampled high-frequency signal (BtmPf and / or BtmPr) from the directional coupler 60. In addition, the abnormality detection device 70 acquires a sampled AE signal (BtmAE) from the AE sensor 61. The abnormality detection device 70 determines the presence / absence of abnormal discharge by using both the result of analyzing the waveform pattern of the high-frequency signal and the result of analyzing the waveform pattern of the AE signal. Accordingly, when only the AE sensor 61 is used, it is difficult to separate the abnormal discharge measurement value and noise included in the AE signal, but the reflected wave or traveling wave high-frequency signal from the directional coupler is difficult. The noise included in the detection result of the AE sensor 61 can be removed using. In the detection of abnormal discharge according to the present embodiment, abnormal discharge is detected using both sampling data detected by the directional coupler 60 and the AE sensor 61. However, the sampling data detected by the AE sensor 61 is not necessarily limited. There is no need to detect abnormal discharge by analyzing only the sampling data detected by the directional coupler 60. However, it is more preferable that the abnormal discharge is detected by using both sampling data because the accuracy is increased.

  The sampling of the high frequency signal is preferably performed every 1 μsec to 5 μsec, and the sampling of the AE signal is preferably performed every 1 μsec to 1 msec.

  If the sampling data of the AE signal is always collected during the plasma processing, the collected data amount becomes enormous. Therefore, analyzing all the collected data and detecting abnormal discharge is inefficient due to the large processing load. On the other hand, if conditions that are likely to cause abnormal discharge can be properly extracted, sampling data is collected only while the conditions are satisfied, and only necessary sampling data needs to be analyzed, so the processing load is low. It is efficient.

(When removing wafer)
In contrast, the inventor has found through experiments that abnormal discharge is likely to occur on the wafer W during static elimination of the wafer. Therefore, in the present embodiment, sampling data is collected according to the timing at the time of wafer detachment where abnormal discharge on the wafer W is likely to occur. Thereby, it is possible to detect a minute abnormal discharge that occurs on the wafer and on the back surface of the wafer. Here, the time of wafer detachment refers to the period from the start of wafer detachment to the end of wafer W detachment. Specifically, after the plasma processing, when the output of the DC high voltage power HV applied to the electrode 13a of the electrostatic chuck 13 is turned off, or when the DC high voltage power HV is reversely applied, the detachment of the wafer after the plasma processing is started. And set the conditions. Further, after the wafer detachment is started, a condition for ending the wafer W detachment is set when the pin for lifting the wafer W mounted on the mounting table 12 is raised and the transfer gate valve 36 is opened.

  As described above, in this embodiment, when the output of the DC high-voltage power HV is turned off, when the DC high-voltage power HV is reversely applied, when the pins are moved up and down, and when the transfer gate valve 36 is opened and closed, the wafer W is used. Since abnormal discharge is likely to occur, sampling data is collected during the wafer removal. However, the timing when the wafer is detached is not limited to this, and it may be from when the output of the DC high-voltage power HV is turned off until when the transfer gate valve 36 is opened or closed, and the output of the DC high-voltage power HV is turned off. It may be from when the pin is moved up and down, or may be from when the DC high voltage power HV is reversely applied to when the transfer gate valve 36 is opened and closed, and when DC high voltage power HV is reversely applied. It may be from when the pin is raised and lowered.

[Function configuration]
Next, the functional configuration of the abnormality detection device 70 will be described with reference to FIG.

  The abnormality detection device 70 is a device that detects abnormal discharge of the etching processing apparatus 10, and includes a plasma processing control unit 71, a monitoring unit 72, an acquisition unit 73, an analysis unit 74, an abnormality determination unit 75, and a storage unit 76. However, the plasma processing control unit 71 and the storage unit 76 are not necessarily functions essential to the abnormality detection device 70.

  The plasma processing control unit 71 controls an etching process performed by the etching processing apparatus 10. Specifically, the plasma processing control unit 71 controls the power and on / off of the high frequency power from the second high frequency power supply 50, and controls the power and on / off of the DC high voltage power HV of the DC high voltage power supply 14. . Further, the plasma processing control unit 71 controls the opening and closing of the transfer gate valve 36, and controls the vertical movement of the pins for transferring the wafer W provided on the mounting table 12 (not shown).

  The monitoring unit 72 monitors the operation of the wafer W placed in the processing chamber C from the start of desorption of the wafer W after the plasma processing until the transfer gate valve 36 is opened, and the operation is monitored. Identified as a release action.

  The acquisition unit 73 is a directional coupler provided between the matching unit 51 and the second high-frequency power source 50 that applies high-frequency power to the processing chamber C during operation when the specified wafer W is detached. A high-frequency signal of at least one of a traveling wave and a reflected wave output from 60 is acquired. The acquisition unit 73 acquires an AE signal output from the AE sensor 61 for detecting acoustic emission (AE) generated in the processing chamber C. A directional coupler is provided between the first high frequency power supply 40 and the matching unit 41, and the acquisition unit 73 receives at least one of the traveling wave and the reflected wave output from the directional coupler. You may get it.

  The analysis unit 74 analyzes the waveform pattern of the acquired high frequency signal. The analysis unit 74 also analyzes the waveform pattern of the acquired AE signal. As an example of the analysis method, the analysis unit 74 determines the maximum amplitude value of the high frequency signal and the maximum amplitude value of the AE signal during operation when the wafer W is detached based on the waveform pattern of the high frequency signal and the waveform pattern of the AE signal. It may be extracted and analyzed. Further, the analysis unit 74 performs frequency analysis (FFT: Fast Fourier Transform) on the waveform pattern of the AE signal, removes noise from the frequency-analyzed data using a desired noise removal filter, and then obtains the data from which the noise has been removed. You may analyze.

  The noise removal filter 62 is a filter used for removing unnecessary noise from the AE signal. The noise removal filter 62 removes noise in a predetermined band. For example, as shown in FIG. 9, the noise removal filter 62 passes only a signal of 70 kHz or higher and passes only a signal of a desired band through a high-pass filter (HPF) that removes other bands as noise. There is a band-pass filter (BPF) that removes the other bands as noise. For example, in this embodiment, the bandpass filter (BPF) band is set to 80 to 105 kHz, 130 to 150 kHz, 220 to 240 kHz, 260 to 290 kHz, 300 to 325 kHz, 430 to 450 kHz, and the other bands are set to noise. Remove as.

  The analysis unit 74 converts the AE signal passed through the noise removal filter 62 into digital data (high-speed sampling data) by high-speed sampling at a frequency of 1 MHz, for example, and stores the digital data in the storage unit 76.

  The abnormality determination unit 75 determines the presence or absence of abnormal discharge based on the analysis result of the waveform pattern of the high-frequency signal. The abnormality determination unit 75 may determine the presence or absence of abnormal discharge based on the analysis result of the waveform pattern of the high-frequency signal and the analysis result of the waveform pattern of the AE signal. Specifically, the abnormality determination unit 75 compares the maximum amplitude of the high-frequency signal with a first threshold (threshold C described later), and compares the maximum amplitude of the AE signal with a second threshold (threshold D described later). By doing so, the presence or absence of abnormal discharge may be determined. Further, in consideration of noise included in the AE signal, the abnormality determination unit 75 determines the presence or absence of abnormal discharge by comparing the generation time of the maximum amplitude of the high frequency signal and the generation time of the maximum amplitude of the AE signal. May be.

  The memory | storage part 76 memorize | stores the various threshold values for the abnormality determination part 75 determining whether it is abnormal discharge. The storage unit 76 may temporarily store sampling data from the AE sensor 61 and sampling data from the directional coupler 60. Various threshold values may be optimized by feeding back abnormal discharge detection results.

  The functions of the plasma processing control unit 71, the monitoring unit 72, the acquisition unit 73, the analysis unit 74, and the abnormality determination unit 75 are performed by, for example, a CPU (Central Processing Unit) operating according to a program stored in the storage unit 76. Can be realized. This program may be provided by being stored in a storage medium and read into the storage unit 76 via a driver (not shown), or may be downloaded from a network (not shown) and stored in the storage unit 76. There may be. Further, a DSP (Digital Signal Processor) may be used instead of the CPU in order to realize the functions of the above-described units. The storage unit 76 can be realized as a RAM (Random Access Memory) or a ROM (Read Only Memory) using, for example, a semiconductor memory, a magnetic disk, or an optical disk. The functions of the plasma processing control unit 71, the monitoring unit 72, the acquisition unit 73, the analysis unit 74, and the abnormality determination unit 75 may be realized by operating using software, and operate using hardware. May be realized.

[Operation: Data acquisition]
Next, the data acquisition operation of the abnormality detection device 70 according to the present embodiment will be described with reference to FIG. FIG. 3 is a flowchart of data acquisition processing executed by the abnormality detection device 70 according to the present embodiment.

  In the data acquisition process, the acquisition unit 73 first determines whether or not the plasma process has been completed (step S305). If the plasma processing has not ended, the acquisition unit 73 repeats step S305 until the plasma processing ends. When the plasma processing is completed, the acquisition unit 73 determines whether the DC high-voltage power HV from the DC high-voltage power supply 14 is turned off and the reverse application of the DC high-voltage power HV is executed (step S310). When the reverse application of the DC high-voltage power HV is not executed, the acquisition unit 73 repeats Step S310 until the reverse application of the DC high-voltage power HV is executed.

  When the reverse application of the DC high-voltage power HV is executed, the acquisition unit 73 accumulates the data of the high frequency signal and the AE signal in the storage unit 76 as sampling data at the time of wafer detachment (Step S315). Next, the acquisition unit 73 determines whether or not the transfer gate valve 36 for unloading the wafer W is opened (step S320). The acquisition unit 73 acquires sampling data until the transfer gate valve 36 is opened, and ends the processing when the transfer gate valve 36 is opened.

  An example of data sampled by the data acquisition process described above is shown in FIG. The horizontal axis of each graph in FIG. 4 is time, and the vertical axis is voltage. Each graph on the left side of FIG. 4 shows, in order from the top, high-frequency signal (BtmPf) data of traveling wave, high-frequency signal (BtmPr) of reflected wave, data of AE signal (BtmAE), and data of DC high-voltage power HV It is.

  The portion where the voltage value sharply decreases in the traveling wave high frequency signal (BtmPf) and the reflected wave high frequency signal (BtmPr) is when the plasma processing is completed, and the high frequency power of the second high frequency power supply 50 is turned off. Show the time.

  Each graph in the center of FIG. 4 is an enlarged version of each graph on the left side of FIG. As shown in the lowermost graph in the center of FIG. 4, the high-frequency signal (BtmPf) of the traveling wave shown in the upper three in FIG. From the high frequency signal (BtmPr) and AE signal (BtmAE) of the reflected wave, it can be seen that an abnormal discharge on the wafer indicated by a thick arrow in the figure occurs. Each graph on the right side of FIG. 4 is an enlarged view of two graphs from the top in the center of FIG. When a defect inspection of the wafer was performed separately, discharge traces were confirmed on the wafer in which an abnormal waveform as shown in FIG. 4 was detected. From the above experimental results, it has been found that there is a correlation between the operation when the wafer is detached and the abnormal discharge on the wafer.

[Operation: Detection of abnormal discharge]
Based on the above examination results, the abnormal discharge detection operation of the abnormality detection device 70 according to the present embodiment will be described below with reference to FIG. FIG. 5 shows a flowchart of an abnormal discharge detection process executed by the abnormality detection device 70 according to the present embodiment.

  In the abnormal discharge detection process, first, the analysis unit 74 acquires the sampling data at the time of wafer detachment acquired by the acquisition unit 73 (step S505). For example, the analysis unit 74 reads sampling data at the time of wafer detachment from the storage unit 76.

  Next, the abnormality determination unit 75 determines whether there is an abnormal peak having a value larger than a predetermined threshold A in the sampling data of the high-frequency signal (step S510). If there is an abnormal peak, the abnormality determination unit 75 determines that there is an abnormal discharge on the wafer W (step S515), and instructs the plasma processing control unit 71 to stop the process execution (step S520). finish.

  On the other hand, in step S510, when there is no abnormal peak having a value larger than the predetermined threshold A among the sampling data of the high frequency signal, the abnormality determining unit 75 determines the predetermined among the sampling data of the AE signal. It is determined whether there is an abnormal peak having a value larger than the threshold value B (step S525). If there is no abnormal peak, this process is terminated. On the other hand, if there is an abnormal peak having a value larger than a predetermined threshold B in the sampling data of the AE signal, the abnormality determining unit 75 determines that there is an abnormal discharge on the wafer W (step S515), and the process After instructing the plasma processing control unit 71 to stop the execution (step S520), the present processing is terminated.

  FIG. 6 shows an example of data relating to the high frequency signal among the data used for the abnormal discharge process described above. The horizontal axis of each graph in FIG. 6 is time, and the vertical axis is voltage. Each graph of FIG. 6 is data of the high-frequency signal (BtmPr) of the reflected wave of the directional coupler 60 when the DC high-voltage power HV is reversely applied to the first to twelfth wafers.

  Looking at the state of the high-frequency signal (BtmPr) of the reflected wave, the second wafer, the fifth wafer, the sixth wafer, the eighth wafer, and the ninth sheet when the DC high voltage power HV is reversely applied. A spectrum suspected of having an abnormal peak was detected in the wafer (arrow portion in FIG. 6).

  Therefore, the abnormality determining unit 75 compares the maximum amplitude of the high-frequency signal with the threshold A (corresponding to the first threshold) and determines the presence or absence of abnormal discharge when performing the comparison process in step S510. FIG. 7 shows a comparison between the maximum amplitude of the high-frequency signal (BtmPr) of the reflected wave and the threshold value A. According to this, the abnormality determination unit 75 determines that abnormal discharge has occurred in the second, fifth, sixth, and ninth wafers, and low-level abnormal discharge has occurred in the eighth wafer. Is determined. When the inventor checked the arcing mark with the SEM (scanning electron microscope), the arcing mark generated on the second, fifth, sixth and ninth wafers was larger than the arcing mark generated on the eighth wafer. It was. It is determined that an abnormal discharge has occurred above the threshold C, and the magnitude of the abnormal discharge is distinguished by the magnitude of the voltage value. That is, it is determined that a relatively large abnormal discharge is greater than or equal to the threshold A, and a low degree abnormal discharge is greater than or equal to the threshold C.

  Thus, it was found that the wafer having an abnormal waveform coincided with the wafer having an arcing mark in the high frequency signal (BtmPr) of the reflected wave when HV was reversely applied. It was also found that there was a correlation between the value of the maximum amplitude of the high-frequency signal (BtmPr) of the reflected wave and the size of the arcing mark.

  In the abnormality detection method according to the present embodiment, control is performed so that the plasma processing is immediately stopped when even one sheet of abnormal discharge is detected. Note that the threshold value C will be described in a later-described modification.

  Next, FIG. 8 shows an example of data related to the AE signal among data used for the abnormal discharge process. The horizontal axis of each graph in FIG. 8 is time, and the vertical axis is voltage. Each graph of FIG. 8 is data of the AE signal (BtmAE) of the AE sensor 61 when the DC high voltage power HV is reversely applied to the first to twelfth wafers.

  Looking at the state of the AE signal (BtmAE), the spectrum suspected of being an abnormal peak is detected in the second, fifth, sixth, eighth, and ninth wafers when the DC high-voltage power HV is reversely applied (arrow in FIG. 8). portion).

  Therefore, when performing the comparison process in step S525, the abnormality determination unit 75 compares the maximum amplitude of the AE signal after noise removal with the threshold B (corresponding to the second threshold), and determines the presence or absence of abnormal discharge. To do. At that time, the analysis unit 74 performs frequency analysis on the AE signal. Each graph on the left side of FIG. 9 is an AE signal for the second, fifth, sixth, eighth, and ninth wafers, and each graph in the center of FIG. 9 is the second, fifth, sixth, eighth, and ninth sheets. It is the data after the frequency analysis of the AE signal about this wafer. The arrows shown in each graph on the left side of FIG. 9 and each graph in the center of FIG.

  Further, the analysis unit 74 passes the data after frequency analysis through the above-described noise removal filter. FIG. 10 shows a comparison between the maximum amplitude of the AE signal (BtmAE) after noise removal and the threshold value B. FIG. According to this, the abnormality determination unit 75 determines that abnormal discharge has occurred in the second, fifth, sixth, and ninth wafers in which the maximum amplitude of the AE signal after noise removal is larger than the threshold value B, and noise removal is performed. It is determined that a low degree of abnormal discharge has occurred in the eighth wafer in which the maximum amplitude of the subsequent AE signal is smaller than the threshold value B. However, in the abnormality detection method according to the present embodiment, when even one sheet detects abnormal discharge, control is performed to immediately stop the plasma processing. It is determined that an abnormal discharge has occurred above the threshold D, and the magnitude of the abnormal discharge is distinguished by the magnitude of the maximum amplitude of BtmAE after noise removal. That is, it is determined that a relatively large abnormal discharge is greater than or equal to the threshold B, and a low degree abnormal discharge is greater than or equal to the threshold D.

  Comparing the above results with the results of wafer defect inspection, it was found that the wafer having an abnormal waveform in the AE signal (BtmAE) and the wafer having an arcing mark coincided with the reverse application of HV. It was also found that there was a correlation between the maximum amplitude of the high frequency signal (BtmPr), the maximum amplitude of the AE signal (BtmAE), and the size of the arcing mark.

  The AE sensor 61 has a property that the sensitivity of the sensor is very good. Therefore, the AE signal detected by the AE sensor 61 is not only a signal due to abnormal plasma discharge but also a machine due to opening / closing of the transfer gate valve 36 of the plasma processing apparatus 10 and the vertical movement of a pin for wafer transfer. It contains a lot of noise such as mechanical vibration. Therefore, there is a concern that the detection accuracy of abnormal plasma discharge may be reduced due to noise included in the AE signal. On the other hand, in the abnormality detection method according to the present embodiment, both the analysis of the high frequency signal and the analysis of the AE signal are analyzed. And when abnormal discharge is detected based on those analysis results, it controls to stop plasma processing immediately. For example, the presence / absence of abnormal discharge can be determined by comparing not only the amplitude of each signal but also the generation time of the maximum amplitude of the high frequency signal and the generation time of the maximum amplitude of the AE signal. Since the AE sensor detects mechanical vibration, the propagation speed of vibration from the abnormal discharge occurrence place to the AE sensor changes depending on the installation place and installation method of the AE sensor. Therefore, for example, as shown in FIG. 11, there is a difference between the occurrence time of the abnormal peak in the high frequency signal and the occurrence time of the abnormal peak in the AE signal.

  When comparing the generation time of the maximum amplitude of the specific high frequency signal and the generation time of the maximum amplitude of the AE signal, first, in step S525 instead of step S510 in FIG. 5, whether there is an abnormal peak in the sampling data of the AE signal. Determine. If there is an abnormal peak, the process advances to step S510 to determine whether there is an abnormal peak in the sampling data of the high-frequency signal. If there is an abnormal peak in step S510, the generation time of the maximum amplitude of the high-frequency signal is compared with the generation time of the maximum amplitude of the AE signal. If the time lag exceeds a predetermined threshold, it is determined that the signal included in the AE signal is noise, and if the time lag falls within the threshold, it is determined that abnormal discharge has occurred. . In this way, by removing the noise included in the AE signal based on the analysis result of the high frequency signal, it is possible to improve the detection accuracy of the abnormal plasma discharge. An abnormal discharge can be detected based only on the analysis result of the high-frequency signal. On the other hand, it is not preferable to detect abnormal discharge based only on the analysis result of the AE signal because the noise cannot be removed.

  As described above, according to the abnormality detection device 70 according to the present embodiment, a minute abnormal discharge can be detected from the sampling data. Such real-time diagnosis can stop abnormal processing at an early stage and can prevent a decrease in yield.

  Further, in the abnormality detection method according to the present embodiment, sampling data is collected according to the timing at the time of wafer detachment where abnormal discharge on the wafer W is likely to occur. According to this, since it is sufficient to analyze only necessary sampling data, it is possible to reduce the processing load and improve efficiency.

  In the above embodiment, the directional coupler 60 is used to detect the high frequency signal, but the method of detecting the high frequency signal is not limited to the contents of the present embodiment. For the detection of the high frequency signal, a so-called RF sensor such as an RF probe for detecting a high frequency voltage or a current probe for detecting a high frequency current can be used. The directional coupler 60 is also included in the RF sensor.

  Even when an RF sensor other than the directional coupler 60 is used, as shown in FIG. 12, the RF sensor 80 is more specifically connected between the matching unit 51 of the second high-frequency power supply 50 and the mounting table 12. The power supply rod 52 may be connected. By installing the RF sensor 80 at a position closer to the mounting table 12 that also serves as the lower electrode DE, more accurate detection of a high-frequency signal can be performed. Even when an RF probe or a current probe is used, the abnormal discharge detection operation in the abnormality detection device 70 is the same as described above.

  An example of data sampled using an RF probe as the RF sensor 80 is shown in FIG. The horizontal axis of each graph in FIG. 13 is time, and the vertical axis is voltage. Each graph in FIG. 13 is obtained by enlarging the time axis of the data by the RF probe, the AE signal (BtmAE), and the data by the RF probe in order from the top.

  As shown in FIG. 13, an example of data sampled using an RF probe as the RF sensor 80 is shown in FIG. The horizontal axis of each graph in FIG. 13 is time, and the vertical axis is voltage. Each graph in FIG. 13 is obtained by enlarging the time axis of the data by the RF probe, the AE signal (BtmAE), and the data by the RF probe in order from the top. When a defect inspection of the wafer on which an abnormal waveform was detected as shown in FIG. 13 was performed separately, discharge traces were confirmed. Therefore, it was confirmed that even if a sensor other than the directional coupling 60 is used as the RF sensor 80, the presence or absence of abnormal discharge can be determined.

[Modification]
Finally, an abnormal discharge detection operation of the abnormality detection device 70 according to the modification of the above embodiment will be described with reference to FIG. FIG. 14 shows a flowchart of an abnormal discharge detection process executed by the abnormality detection device 70 according to a modification of the above embodiment.

  In the abnormal discharge detection process according to the modified example, first, the analysis unit 74 acquires the sampling data at the time of wafer detachment acquired by the acquisition unit 73 (step S505). Next, the abnormality determination unit 75 determines whether there is an abnormal peak having a value larger than a predetermined threshold C (see FIG. 7) in the sampling data of the high-frequency signal (step S510). If there is an abnormal peak, the abnormality determination unit 75 determines that there is an abnormal discharge on the wafer W (step S515), and instructs the plasma processing control unit 71 to stop the process execution (step S520). finish.

  On the other hand, in step S510, when there is no abnormal peak having a value larger than the predetermined threshold C among the sampling data of the high frequency signal, the abnormality determination unit 75 determines the predetermined among the sampling data of the AE signal. It is determined whether there is an abnormal peak having a value larger than the threshold value D (step S525). If there is an abnormal peak, the abnormality determination unit 75 determines that there is an abnormal discharge on the wafer W (step S515), and instructs the plasma processing control unit 71 to stop the process execution (step S520). finish.

  As described above, according to the present modification, abnormal discharge occurs when the threshold values C and D in FIGS. 7 and 10 are exceeded. Below threshold A and B, low-level abnormal discharge, and above threshold A and B, distinguish between relatively large-scale abnormal discharge and abnormal discharge, and investigate the cause of abnormal discharge as FDC (Fault Detection and Classification). And lead to countermeasures.

  However, even a low abnormal discharge leads to a decrease in yield, so that it is necessary to stop the plasma processing apparatus by recognizing that “abnormal discharge has occurred” if the thresholds A and B are equal to or higher than the thresholds C and D.

  The preferred embodiments of the present invention have been described in detail above with reference to the accompanying drawings, but the present invention is not limited to such examples. It is obvious that a person having ordinary knowledge in the technical field to which the present invention pertains can make various changes or modifications within the scope of the technical idea described in the claims. Of course, it is understood that these also belong to the technical scope of the present invention.

  For example, the abnormality detection apparatus according to the above-described embodiment has been described by using an example of an etching processing apparatus. However, the abnormality detection apparatus according to the present invention is not limited thereto, and may be, for example, a film forming apparatus, an ashing processing apparatus, or the like. Can be used. As the plasma source of the plasma processing apparatus, various plasmas such as microwave plasma, magnetron plasma, and ICP plasma can be used in addition to the parallel plate plasma in the above-described embodiment.

DESCRIPTION OF SYMBOLS 10 Etching apparatus 12 Mounting stand 13 Electrostatic chuck 14 DC high voltage power supply 36 Transfer gate valve 40 1st high frequency power supply 41 Matching device 50 2nd high frequency power supply 51 Matching device 52 Feeding rod 53a, 53b Coaxial cable 60 Directional coupler 61 AE sensor 62 Noise removal filter 70 Abnormality detection device 71 Plasma processing control unit 72 Monitoring unit 73 Acquisition unit 74 Analysis unit 75 Abnormality determination unit 76 Storage unit 80 RF sensor C Processing chamber UE Upper electrode DE Lower electrode

Claims (15)

A high-frequency signal output from an RF sensor provided between a matching unit of a high-frequency power source that applies high-frequency power to a processing chamber for plasma-processing the object to be processed and a lower electrode that functions as a mounting table for mounting the object to be processed An acquisition unit for acquiring an AE signal output from an AE sensor for detecting acoustic emission generated in the processing chamber;
An analysis unit for analyzing the waveform pattern of the acquired high-frequency signal and the waveform pattern of the AE signal;
An abnormality detection device comprising: an abnormality determination unit that determines presence or absence of abnormal discharge based on an analysis result of the waveform pattern of the high-frequency signal and an analysis result of the waveform pattern of the AE signal. The analysis unit extracts a maximum amplitude value of the high frequency signal and a maximum amplitude value of the AE signal during operation when the object is detached based on the waveform pattern of the high frequency signal and the waveform pattern of the AE signal. ,
The abnormality determination unit determines the presence or absence of abnormal discharge by comparing the maximum amplitude of the high-frequency signal with a first threshold and comparing the maximum amplitude of the AE signal with a second threshold. The abnormality detection device according to claim 1.   The abnormality detection apparatus according to claim 1, wherein the RF sensor is any one of a directional coupler, an RF probe, and a current probe. Monitors the operation of the target object placed in the processing chamber from the start of desorption of the target object after plasma processing until the transfer gate valve opens, and identifies the operation as the action when the target object is released A monitoring unit to
The lower electrode functioning as a mounting table for mounting the object to be processed between the high-frequency power source for applying high-frequency power to the processing chamber and the matching unit during the operation when the object to be processed is detached and the matching unit An acquisition unit for acquiring a high-frequency signal of at least one of a traveling wave and a reflected wave output from a directional coupler provided between
An analysis unit for analyzing a waveform pattern of the acquired high-frequency signal;
An abnormality detection unit comprising: an abnormality determination unit that determines presence or absence of abnormal discharge based on an analysis result of a waveform pattern of the high-frequency signal. The acquisition unit acquires an AE signal output from an AE sensor for detecting acoustic emission (AE) generated in the processing chamber,
The analysis unit analyzes a waveform pattern of the acquired AE signal,
The abnormality detection device according to claim 4, wherein the abnormality determination unit determines presence / absence of abnormal discharge based on an analysis result of a waveform pattern of the high-frequency signal and a waveform pattern of the AE signal. The analysis unit extracts a maximum amplitude value of the high frequency signal and a maximum amplitude value of the AE signal during operation when the object is detached based on the waveform pattern of the high frequency signal and the waveform pattern of the AE signal. ,
The abnormality determination unit determines the presence or absence of abnormal discharge by comparing the maximum amplitude of the high-frequency signal with a first threshold and comparing the maximum amplitude of the AE signal with a second threshold. The abnormality detection device according to claim 5.   The abnormality determination unit determines presence / absence of abnormal discharge by comparing the generation time of the maximum amplitude of the high-frequency signal with the generation time of the maximum amplitude of the AE signal. The abnormality detection apparatus in any one.   The analysis unit performs frequency analysis on the waveform pattern of the AE signal, removes noise from the frequency-analyzed data using a desired noise removal filter, and then analyzes the data from which the noise has been removed. The abnormality detection apparatus as described in any one of 1-3, 5-7.   The said AE sensor is attached to the electric power feeding rod which supplies a high frequency electric power to the lower electrode which functions also as a mounting base which mounts a to-be-processed object, The any one of Claims 1-3 and 5-8 characterized by the above-mentioned. The abnormality detection device described in 1.   The abnormality detection apparatus according to claim 1, wherein the sampling of the high-frequency signal is performed every 1 μsec to 5 μsec.   The abnormality detection apparatus according to claim 1, wherein sampling of the AE signal is performed every 1 μsec to 1 msec.   When the output of the DC high-voltage power applied to the electrode of the electrostatic chuck is turned off, or when the DC high-voltage power is reversely applied, the monitoring unit determines that the object to be processed has started to be detached after the plasma processing. The abnormality detection device according to any one of claims 1 to 11, wherein A high-frequency signal output from an RF sensor provided between a matching unit of a high-frequency power source that applies high-frequency power to a processing chamber for plasma-processing the object to be processed and a lower electrode that functions as a mounting table for mounting the object to be processed And obtaining an AE signal output from an AE sensor for detecting acoustic emission generated in the processing chamber;
Analyzing the waveform pattern of the acquired high-frequency signal and the waveform pattern of the AE signal;
And determining whether or not there is abnormal discharge based on the analysis result of the waveform pattern of the high-frequency signal and the analysis result of the waveform pattern of the AE signal. Monitors the operation of the target object placed in the processing chamber from the start of desorption of the target object after plasma processing until the transfer gate valve opens, and identifies the operation as the action when the target object is released And steps to
The lower electrode functioning as a mounting table for mounting the object to be processed between the high-frequency power source for applying high-frequency power to the processing chamber and the matching unit during the operation when the object to be processed is detached and the matching unit Obtaining a high-frequency signal of at least one of a traveling wave and a reflected wave output from a directional coupler provided between
Analyzing the waveform pattern of the acquired high-frequency signal;
Determining whether or not there is abnormal discharge based on the analysis result of the waveform pattern of the high-frequency signal. A plasma processing apparatus,
A processing chamber for processing the substrate;
Plasma generating means for generating plasma in the processing chamber;
An abnormal discharge detection device connected to a power supply unit of the plasma generation means and detecting an abnormality of the plasma,
The abnormal discharge detection device includes:
A high-frequency signal output from an RF sensor provided between a matching unit of a high-frequency power source that applies high-frequency power to a processing chamber for plasma-processing the object to be processed and a lower electrode that functions as a mounting table for mounting the object to be processed An acquisition unit for acquiring an AE signal output from an AE sensor for detecting acoustic emission generated in the processing chamber;
An analysis unit for analyzing the waveform pattern of the acquired high-frequency signal and the waveform pattern of the AE signal;
A plasma processing apparatus, comprising: an abnormality determination unit that determines presence / absence of abnormal discharge based on an analysis result of a waveform pattern of the high-frequency signal and an analysis result of a waveform pattern of the AE signal.

Images